Dense pedestrian face detection in complex environments

The existing face detection methods were affected by the network model structure used. Most of the face recognition methods had low recognition rate of face key point features due to many parameters and large amount of calculation. In order to improve the recognition accuracy and detection speed of face key points, a real-time face key point detection algorithm based on attention mechanism was proposed in this paper. Due to the multiscale characteristics of face key point features, the deep convolution network model was adopted, the attention module was added to the VGG network structure, the feature enhancement module and feature fusion module were combined to improve the shallow feature representation ability of VGG, and the cascade attention mechanism was used to improve the deep feature representation ability. Experiments showed that the proposed algorithm not only can effectively realize face key point recognition but also has better recognition accuracy and detection speed than other similar methods. This method can provide some theoretical basis and technical support for face detection in complex environment.

Now the epidemic prevention and control has been continuing, frequent removal and wearing of masks in railway stations, subway stations and places such as commuting to and from work are prone to the spread of the virus, and face detection is an important part of the face recognition system. Aiming at the complex environmental factors such as partial occlusion, angle change, light intensity and face blur in face detection in these places, This paper improves the detection accuracy by improving the RetinaFace algorithm. Firstly, the lightweight GhostNet network is introduced to substitute the former MobileNet0.25 backbone network of RetinaFace, and a lightweight model improved version of RetinaFace is obtained, which not only ensures that the model is smaller but also ensures the speed of face detection; In addition, The efficient ECA channel attention mechanism is fused in the enhanced feature extraction network of the model to further enhance the detection performance of small face samples in complex environments. Finally, the simulation conclusion show that compared with the former RetinaFace algorithm, the detection performance of this method in the verification set of different levels of the reconstructed WIDER FACE dataset reaches 93.4% (Easy), 90.8% (Medium) and 77.1% (Hard), which is improved by 2.7 percentage points, 2.2 percentage points and 5 percentage points, respectively. It can be seen that after the introduction of GhostNet network and ECA attention mechanism, the recognition accuracy of faces in complex environments is further improved and network performance is improved.

Prohibited item detection is crucial for the safety of public places. Deep learning, one of the mainstream methods in prohibited item detection tasks, has shown superior performance far beyond traditional prohibited item detection methods. However, most neural network architectures in deep learning still lack sufficient local feature representation ability for overlapping and small targets, and ignore the problem of semantic conflicts caused by direct feature fusion. In this paper, we propose MFPIDet, a novel prohibited item detection neural network architecture based on improved YOLOV7 to achieve reliable prohibited item detection in complex environments. Specifically, a multi-scale attention module (MAM) backbone is proposed to filter the redundant information of target regions and further applied to enhance the local feature representation ability of overlapping objects. Here, to reduce the redundant information of target regions, a squeeze-excitation (SE) block is used to filter the background. Then, aiming at enhancing the feature expression ability of overlapping objects, a multi-scale feature extraction module (MFEM) is designed for local feature representation. In addition, to obtain richer context information, We design an adaptive fusion feature pyramid network (AF-FPN) to combine the adaptive context information fusion module (ACIFM) with the feature fusion module (FFM) to improve the neck structure of YOLOV7. The proposed method is validated on the PIDray dataset, and the tested results showed that our method obtained the highest mAP (68.7%), which is improved by 3.5% than YOLOV7 methods. Our approach provides a new design pattern for prohibited item detection in complex environments and shows the development potential of deep learning in related fields.

Face detection in complex environments is an unsolved problem which has fundamental importance to face recognition, model based video coding, content based image retrieval, and human computer interaction. In this paper we model the face detection problem using information theory, and formulate information based measures for detecting faces by maximizing the feature class separation. The underlying principle is that search through an image can be viewed as a reduction of uncertainty in the classification of the image. The face detection algorithm is empirically compared using multiple test sets, which include four face databases from three universities.

S-AdaBoost is a new variant of AdaBoost and is more effective than the conventional AdaBoost in handling outliers in pattern detection and classification in real world complex environment. Utilizing the divide and conquer principle, S-AdaBoost divides the input space into a few sub-spaces and uses dedicated classifiers to classify patterns in the sub-spaces. The final classification result is the combination of the outputs of the dedicated classifiers. S-AdaBoost system is made up of an AdaBoost divider, an AdaBoost classifier, a dedicated classifier for outliers, and a non-linear combiner. In addition to presenting face detection test results in a complex airport environment, we have also conducted experiments on a number of benchmark databases to test the algorithm. The experiment results clearly show S-AdaBoost's effectiveness in pattern detection and classification.

Empirical analysis of cascade deformable models for multi-view face detection

The depth face detector based on an anchor frame has demonstrated good performance with the advancement of science and technology, but tiny faces and partially occluded faces are still difficult to identify under the effect of posture, lighting, occlusion, and other factors. We present the ESSFD, an enhanced single-stage face detector, to increase the robustness of face detection in complex environments. The following are the important points: (1) Enhancing feature extraction by using an optimized ResNet-D network in the backbone network and adding Atrous Convolution Layer supplementary feature pyramids; (2) For data augmentation, use a combination of Gaussian blur and color jitter to lessen the impact of image environmental elements and increase the model’s robustness; (3) Fine-tune the training parameters for the model. In the presence of unusual postures, complicated lighting, and partial occlusion, ESSFD can increase the detection accuracy of tiny faces when compared to RetinaFace. Experiments reveal that in the easy and medium stages of the WIDER FACE dataset, the Average Accuracy (AP) of the ESSFD detector is around 1% higher than that of the advanced RetinaFace face detector (AP=95.373%). In the hard tiny face detection stage, it is around 2% higher than RetinaFace.

Under the field of object detection, face detection is the fundamental element of computer vision technology, for a few decades it become the popular research area. However due to the limitation of some scenes image data have low resolution, occlusion, illumination, makeup and others issue that make face detection hard. Today many recent algorithms can recognize face detection well under the condition of positive and high resolution. However most of the faces were located in challenged condition and low resolution are the problem of face detection. Small Scale Face Detection is one of the challenged faces that hard to detect easily. Because of the low resolution, scale invariance and contextual reasoning, due to the lack of detail information about located face. To achieve the required goal of small scale face detection, this research work has proposed a novel Small Scale Targeted Face Detection (SSTFD) and the combination of MS-faster-RCNN to increase the accuracy of multi scale face. The proposed approach has used Residual Network (ResNet) 50, 101 and 152 as the backbone network and from Intersection over Union (IoU) regression, the Complete Intersection over Union (CIoU) and Control Distance Intersection over Union (CDIoU) are used. Also, this research work has applied various techniques like small scale targeted face detection to produce high accuracy results and good speed.

Face detection and recognition plays an important role in many occasions. This study explored the application of convolutional neural network in face detection and recognition. Firstly, convolutional neural network was briefly analyzed, and then a face detection model including three convolution layers, four pooling layers, introduction layers and three fully connected layers was designed. In face recognition, the self-learning convolutional neural network (CNN) model for global and local extended learning and Spatial Pyramid Pooling (SPP)-NET model were established. LFW data sets were used as model test samples. The results showed that the face detection model had an accuracy rate of 99%. In face recognition, the self-learning CNN model had an accuracy rate of 94.9% accuracy, and the SPP-Net model had an accuracy rate of 92.85%. It suggests that the face detection and recognition model based on convolutional neural network has good accuracy, and the face recognition efficiency of self-learning CNN model was better, which deserves further research and promotion.

With the development of depth learning and synthetic aperture radar (Synthetic Aperture Radar, SAR) technology, SAR image target detection based on convolution neural network (convolutional neural network, CNN) has achieved certain results. However, there are still problems in SAR detection of near-shore ship targets in complex environments. For improving the detection performance of the algorithm, the detection rate of SAR image near shore ship targets in complex environment is improved. This paper proposes an algorithm for SAR image ship target detection in complex environment. The algorithm first uses convolution neural network for coastal segmentation, and SAR image ship target detection through the results of coastal segmentation. The experimental results show that the algorithm has efficient detection ability for SAR image near-shore ship target detection in complex environment.

During the detection in complex biological system, similar to other fluorescence probes, the molecular beacons suffer severely from the background signal interference. Recent studies indicated that excimer molecular beacon (EMB) can address the problem. EMB is a dual-pyrene-labeled hairpin DNA structure with large Stokes shift and long fluorescence lifetime, which afford an effective strategy for detection in complex biological environment. In the chapter, the recent development of the research of EMB is presented. Firstly, the general design of EMB, as well as the structure and working mechanism, is introduced. Furthermore, the synthesis and properties of EMB are descripted, which explain its capability of detection in complex environment and high sensitivity and selectivity. Finally, the examples of the different applications are discussed, including nucleic acids and other molecules detection.

This paper presents an improved face detection method for color images. We propose a boosted skin-color model in RGB space which can reduce more effectively noises forming from similar skin colors. With our solution, we receive more reasonable skin detection for different human races. We modifed local binary pattern (LBP) by adding a set of spatial templates. This LBP considers both principal local shapes and spatial textures of facial components. Human face is represented by LBP histogram. Moreover, the grayscale image of human face is changed to discrete cosine transform (DCT) coefficients used in embedded hidden Markov models (eHMMs). A modified LBP (mLBP) histogram matching and eHMMs are composed to hierarchical classifier to determine whether skin regions are faces or not. The experiments show that our method performs a better capability for face detection in complex environments than using separately eHMMs or LBP histogram. The correct face detection rate of proposed system is over 94% among our test database which consists totally 485 single and multi-face color images of 1429 persons in different lighting conditions, face rotations, occlusions and complex backgrounds from different sources: Caltech face database, Sumgmug image library, family photos, personal digital images and World Wide Web

ABSTRACTIn recent years, the development of deep learning algorithms has significantly advanced the application of synthetic aperture radar (SAR) aircraft detection in remote sensing and military fields. However, existing methods face a dual dilemma: CNN‐based models suffer from insufficient detection accuracy due to limitations in local receptive fields, whereas Transformer‐based models improve accuracy by leveraging attention mechanisms but incur significant computational overhead due to their quadratic complexity. This imbalance between accuracy and efficiency severely limits the development of SAR aircraft detection. To address this problem, this paper propose a novel neural network based on state space models (SSM), termed the Mamba SAR detection network (MSAD). Specifically, we design a feature encoding module, MEBlock, that integrates CNN with SSM to enhance global feature modelling capabilities. Meanwhile, the linear computational complexity brought by SSM is superior to that of Transformer architectures, achieving a reduction in computational overhead. Additionally, we propose a context‐aware feature fusion module (CAFF) that combines attention mechanisms to achieve adaptive fusion of multi‐scale features. Lastly, a lightweight parameter‐shared detection head (PSHead) is utilised to effectively reduce redundant parameters through implicit feature interaction. Experiments on the SAR‐AirCraft‐v1.0 and SADD datasets show that MSAD achieves higher accuracy than existing algorithms, whereas its GFLOPs are 2.7 times smaller than those of the Transformer architecture RT‐DETR. These results validate the core role of SSM as an accuracy‐efficiency balancer, reflecting MSAD's perceptual capability and performance in SAR aircraft detection in complex environments.

Face mask-wearing detection is of great significance for safety protection during the epidemic. Aiming at the problem of low detection accuracy due to the problems of occlusion, complex illumination, and density in mask-wearing detection, this paper proposes a neural network model based on the loss function and attention mechanism for mask-wearing detection in complex environments. Based on YOLOv5s, we first introduce an attention mechanism in the feature fusion process to improve feature utilization, study the effect of different attention mechanisms (CBAM, SE, and CA) on improving deep network models, and then explore the influence of different bounding box loss functions (GIoU, CIoU, and DIoU) on mask-wearing recognition. CIoU is used as the frame regression loss function to improve the positioning accuracy. By collecting 7,958 mask-wearing images and a large number of images of people without masks as a dataset and using YOLOv5s as the benchmark model, the mAP of the model proposed in the paper reached 90.96% on the validation set, which is significantly better than the traditional deep learning method. Mask-wearing detection is carried out in a real environment, and the experimental results of the proposed method can meet the daily detection requirements.

For the problem of environmental sound detection in complex environments, the environmental sound detection method based on acoustic multidimensional synergistic features and MobileNet-EAL is proposed. Firstly, the acoustic multidimensional synergistic features of the input audios are extracted to achieve the complementary frequency and time domain information, the complementary static and dynamic information to enhance the ability to express the temporal characteristics of the sound signal. Secondly, the ECA_Net channel attention mechanism is added to the MobileNet-EAL network model proposed to make the model ignore the influence and interference caused by the background sound. Meanwhile, the Leaky ReLU activation function is also used to enhance the model's extraction of negative audio feature information to achieve sound detection in complex environments. The experimental results show that the classification accuracy reaches 91.4% on the UrbanSound8K dataset. Compared with traditional environmental sound recognition methods, the method adopted in this paper has obvious advantages in recognition rate and recognition efficiency.